Download recurrent erysipelas despite antibiotic prophylaxis: an

Re v i e w a r t i c l e
Recurrent erysipelas despite antibiotic
prophylaxis: an analysis from case studies
J.B. Koster, B.J. Kullberg, J.W.M. van der Meer*
Department of General Internal Medicine, Radboud University Nijmegen Medical Centre,
and Nijmegen University Centre for Infectious Diseases, PO Box 9101, 6500 HB Nijmegen,
the Netherlands, *corresponding author: tel.: +31 (0)24-361 88 19, e-mail: [email protected]
ABSTR ACT
I N T RODU C T IO N
Background: Erysipelas is a distinctive type of superficial
cellulitis of the skin with prominent lymphatic involvement,
generally caused by group A streptococci. A substantial
proportion of patients experience recurrences of erysipelas,
and this may be a reason to install prophylactic antibiotic
treatment. Despite such prophylaxis, further recurrences
are occasionally encountered.
Objectives: To investigate recurrences of erysipelas during
prophylactic antibiotic treatment and to delineate the
reasons for such failure.
Methods: Retrospective chart review of 117 adult patients
with episodes of erysipelas known in our institution
between 1990 and 2004.
Results: Recurrent episodes of erysipelas, despite
prophylactic treatment, were found in eight patients. Our
analysis indicated noncompliance, incorrect selection and
insufficient dosing of antibiotics, and causative pathogens
other than streptococci as demonstrable causes of the
recurrence of erysipelas. In three patients, a reason for
failure could not be identified.
Conclusions: In a minority of cases, erysipelas recurs
despite antibiotic prophylaxis. Based on these cases, we
first recommend that all efforts are made to (re)confirm the
diagnosis of erysipelas and search for the causative microorganism. Based on this information, the right antibiotic
with adequate dosing and timing can be selected. The issue
of compliance with the prophylactic treatment should be
addressed and finally, the clinician should be aware that
prophylaxis does not prevent erysipelas in all cases.
Erysipelas is an acute inflammation of the skin, with
marked involvement of cutaneous lymphatic vessels. It
is a clinically recognisable entity, with sudden onset of
fever and a painful erythematous swollen lesion, sharply
demarcated from the normal skin. Erysipelas is most
commonly caused by b-haemolytic streptococci of group
A, less so by group B, C, or G streptococci, and occasionally
by Staphylococcus aureus.1,2 In many patients, factors are
present that facilitate the development of erysipelas. The
major risk factors in erysipelas are disruption of the skin
and lymphoedema.3 Erysipelas can be treated successfully
with narrow-spectrum penicillins, such as benzylpenicillin
and feneticillin. Patients who are allergic to penicillin may
be treated effectively with macrolides.
The recurrence rate of erysipelas is high: nearly 30% has
been noted within a two to four year period. 4 Unfortunately,
once erysipelas occurs, damage to the cutaneous lymph
vessels often leads to susceptibility for further relapses.
To prevent a vicious circle of recurrent erysipelas and
vulnerability to subsequent episodes, long-term antibiotic
prophylaxis is advocated. Studies have suggested an effect
of such prophylaxis, using various antibiotic regimens. 4-7
A neglected aspect in studies on prophylaxis for erysipelas
is that a few patients still have attacks of erysipelas during
antibiotic prophylaxis.7,8 In line with this notion, we have
encountered patients with recurrent erysipelas despite
antibiotic prophylaxis. In this paper, we review these cases
and examine the reasons for failure of prophylaxis.
P A T I E N T S A N D M E T HODS
K E Y WORDS
Patients with one or more episodes of erysipelas during
antibiotic prophylaxis were identified and retrospectively
analysed to examine the incidence of recurring erysipelas
Antibiotic prophylaxis, erysipelas, prevention, recurrent
erysipelas
© 2007 Van Zuiden Communications B.V. All rights reserved.
MARCH 2007, Vol. 65, No. 3
89
despite prophylactic antibiotic treatment and the factors
associated with this failure. Patients were identified by
searching two databases for the clinical diagnosis of
erysipelas: the first database contains data on infectious
disease consultations performed in our university medical
centre between 1990 and 2004; the second contains the
diagnoses of patients at the outpatient clinic for internal
medicine from 1999 until 2003. In addition, infectious
disease specialists were asked whether they were aware of
additional patients.
From the charts of patients with attacks of erysipelas
during antibiotic prophylaxis, the following data were
collected: 1) personal characteristics: gender, age, weight;
2) underlying diseases and/or factors predisposing for
erysipelas; 3) attacks of erysipelas before prophylactic
treatment; 4) results of diagnostic tests, such as cultures,
antistreptolysin titres and anti-DNAse B; 5) treatment of
the episodes of erysipelas; 6) prophylactic regimens; 7) the
effect of the antibiotic prophylaxis.
He had been treated with benzathine penicillin 1.2 MU
intramuscularly every four weeks as prophylaxis. Despite
this treatment, frequent attacks of erysipelas recurred. It
turned out that relapses occurred shortly before the next
injection was planned.
At the age of 20, the patient was hospitalised for another
episode of erysipelas, which was treated with penicillin
G. On this occasion, interdigital mycosis was found as a
potential portal of entry. In an attempt to reduce the risk of
another attack, antimycotic treatment and elastic stockings
were prescribed, but despite this, episodes of erysipelas
still recurred. The attacks responded well to therapy with
roxitromycin.
The patient was first seen at our department at the age of
22. He was put on a prophylactic regimen of benzathine
penicillin 1.2 MU every two weeks and this preventive
treatment was successfully continued for two years. Six
months after stopping, a new episode occurred, and the
two-weekly preventive regimen was reinstituted. Despite the
prophylaxis, a new attack occurred three months later. This
episode probably occurred because the penicillin injection
was delayed until 3.5 weeks after the previous dose.
The next episode of erysipelas developed 16 days after
the injection of benzathine penicillin. After treatment, a
prophylactic regimen of injections strictly administered
every two weeks was installed. This prevented recurrences
for 2.5 years. On the patient’s request, the frequency of
administration was again reduced to every three weeks
and this led to new episodes of erysipelas. Antibiotic
prophylaxis every two weeks has prevented further attacks
of erysipelas, for one year of follow-up.
R E SUL T S
In this study, 117 patients with erysipelas were identified.
Five patients had an attack despite antibiotic prophylaxis.
Three more patients were retrieved from infectious
diseases specialists. The characteristics of these eight
patients, the sites of erysipelas, and underlying conditions
are provided in table 1.
The phenomenon of recurrent erysipelas under antibiotic
prophylaxis will further be described on the basis of three
illustrative cases.
Patient E
This 24-year-old female experienced a minor trauma of
her chin at the age of five. The first episode of erysipelas
at the age of 14 affected her chin and lower lip. During the
following years she had several attacks of erysipelas. After
her second hospitalisation at the age of 18, she received
Patient C
This 29-year-old male had suffered from several episodes
of erysipelas affecting the right leg since the age of 17. At
15 years, he underwent an epiphysiodesis of the femur
and tibia of his right leg to correct a difference in length.
Table 1. Patients characteristics
Patient
Age
(years)
Sex
Weight
(kg)
Location
Underlying condition
Site of entry
A
45
m
113
Right and left leg
Trauma: spinal cord lesion,
multiple fractures left leg
Intertrigo
B
56
m
–
Left leg
Spinal muscular atrophy,
fracture left femur
Epiphysiodesis right leg
Post-breast cancer surgery
Trauma chin
Short-bowel syndrome, arterial
insufficiency right leg
Post-breast cancer surgery
Recurrent herpes simplex
C
D
E
F
29
53
24
38
m
f
f
m
108
86
72
74
Right leg
Left arm
Face
Right and left leg
G
H
55
39
f
f
62
63
Right arm
Face, neck
47
52
Intertrigo
17
44
14
34
19
47
18
35
Skin lesion
Herpetic lesions
45
29
53
30
Dermatomycosis
Eczema
Koster, et al. Recurrent erysipelas despite antibiotic prophylaxis.
MARCH 2007, Vol. 65, No. 3
90
Age at first Age at start of
episode
prophylaxis
(years)
(years)
33
39
prophylactic treatment, consisting of benzathine penicillin
1.2 MU intramuscularly every four weeks.
She remained free from episodes of erysipelas for 11 months
until a new attack at the same site occurred during antibiotic
prophylaxis. During this episode, the serology was suggestive
for a streptococcal origin: aspartate aminotransaminase
(AST) (332 U ml-1) and anti-DNAse (B 879 U ml-1) were both
elevated, whereas the antistaphylolysin titre was 0.71 U ml-1.
She was referred to our clinic.
The prophylactic regimen was changed to clindamycin
300 mg three times a day. Five months after starting with
this regimen, another episode of erysipelas occurred. As
the patient was not critically ill, we felt confident to try
and see whether it was a dose or a resorption problem. So
we switched to intravenous clindamycin 3 x 600 mg. The
response to therapy with intravenous clindamycin was
rapid. Hereafter, the dose of prophylactic treatment was
raised to 300 mg four times a day. Measurement of serum
clindamycin showed concentrations of 1.69 and 4.12 mg
l-1, which are considered appropriate. Even with the higher
oral dose of clindamycin, another episode of erysipelas
occurred three months later. Prophylaxis was stopped, and
nine months later, she again developed erysipelas.
decreased to 600 mg once daily. This treatment prevented
further episodes of erysipelas for three months, but
thereafter breakthroughs occurred.
An overview of the cases of recurrent erysipelas during
antibiotic prophylaxis is provided in table 2. In most cases,
it is not possible to judge whether these recurrent episodes
represent relapses from foci within the body or exogenous
reinfections.
In a number of cases, a plausible explanation for the failure
of prophylaxis and subsequent recurrences could be given
on the basis of chart review. These were: 1) noncompliance;
2) incorrect antibiotic; 3) other causative micro-organism;
and 4) insufficient antibiotic concentration.
Noncompliance
Two patients experienced an episode of erysipelas when
they extended the interval time of the prophylactic regimes.
Patient C exceeded the prophylactic schedule by 1.5 weeks,
when erysipelas recurred. Patient A had a new attack of
erysipelas ten weeks after the last injection.
Incorrect antibiotic
Instead of benzathine penicillin, patient B received a
combination of benzathine and procaine-benzylpenicillin.
This agent only contains 0.6 MU of benzathine penicillin.
Penicillin concentrations in serum are therefore only
detectable for about one week.
Patient G
This 55-year-old female was healthy until the age of 44,
when breast cancer was diagnosed. A modified radical
mastectomy with axillary lymph node dissection was
performed. A year later, a tumour was found in her
other breast, and a mastectomy with axillary lymph
node dissection was performed followed by radiotherapy.
Tamoxifen was continued for 2.5 years.
Since the second mastectomy, she experienced several
episodes of erysipelas. These episodes would usually
follow a small skin defect on the right hand and were
characterised by a fiery red, sharply demarcated lesion
on one side of the thorax and the right arm, high fever,
and systemic toxicity. Response to treatment with either
amoxicillin or flucloxacillin was rapid.
Five years after surgery, a reconstruction by a latissimus
dorsi muscle flap combined with breast implants was
performed, complicated by oedema of the right arm.
The episodes of erysipelas increased in frequency. After
multiple attacks, the patient was put on benzathine
penicillin 1.2 MU intramuscularly every four weeks. The
attacks decreased but were not completely prevented.
Even injections given every two weeks did not prevent
the episodes of erysipelas. There was no clear correlation
between the time of injection and the occurrence of
erysipelas.
After stopping the prophylactic treatment, the frequency
of erysipelas increased to once every one or two months.
She was put on a prophylactic regimen of clindamycin
600 mg twice daily, and several months later the dose was
Other causative micro-organism
Although no culture samples could be obtained, it is likely
that the recurrences in patients D and B during antibiotic
prophylaxis were not caused by group A streptococci, but by
Staphylococcus aureus. Patient D developed a skin infection
of the left arm four days after the injection of benzathine
penicillin. The infection responded to flucloxacillin. Patient
B developed an attack of erysipelas one week after the
injection of benzathine penicillin 1.2 MU. The initial
treatment consisted of feneticillin 1000 mg four times
a day. Despite an initial improvement, treatment was
unsuccessful. After two weeks, the treatment was changed
to clindamycin 600 mg three times a day, which led to an
improvement.
Insufficient antibiotic concentration
In three patients, the serum concentrations of the
antibiotic agent were probably insufficient. Patient B
received clarithromycin 250 mg daily to prevent further
episodes of erysipelas. This did not prevent the episode
1.5 weeks later. The recommended therapeutic dose for
an adult should be at least 250 mg twice daily. Patients C
and H experienced several episodes during prophylaxis
with benzathine penicillin 1.2 MU. The occurrence of
episodes of erysipelas occurred just before the dose was to
Koster, et al. Recurrent erysipelas despite antibiotic prophylaxis.
MARCH 2007, Vol. 65, No. 3
91
Table 2. Recurrences during antibiotic prophylaxis
Patient
Antibiotic
Recurrence
A
1
B
1
2
3
C
1
2
3
4
D
1
2
E
1
2
3
F
1
G
1
2
H
1
benzathine
penicillin
procaine
penicillin
Dosage
Time to
Treatment of
recurrence erysipelas
penicillin,
feneticillin
Non­­
compliance
Reasons for failure of prophylaxis
Incorrect
Other
Insufficient
No
antibiotic causative
antibiotic explanamicroconcen­
tion
organism
tration
1.2 MU/
4 wk
1.2 MU/
4 wk
15
mo
v
clarithromycin 250
mg/day
benzathine
1.2 MU/
penicillin
3 wk
benzathine
1.2 MU/
penicillin
4 wk
benzathine
1.2 MU/
penicillin
2 wk
benzathine
1.2 MU/
penicillin
2 wk
benzathine
1.2 MU/
penicillin
3 wk
benzathine
1.2 MU/
penicillin
4 wk
benzathine
1.2 MU/
penicillin
3 wk
benzathine
1.2 MU/
penicillin
4 wk
clindamycin
300 mg
t.i.d.
clindamycin
300 mg
q.i.d.
benzathine
1.2 MU/
penicillin
4 wk
benzathine
1.2 MU/
penicillin
4 wk
benzathine
1.2 MU/
penicillin
2 wk
benzathine
1.2 MU/
penicillin
4 wk
1.5
wk
1
wk
*
feneticillin
failed
3.5
wk
2
wk
feneticillin
v
v
2
wk
penicillin,
feneticillin
flucloxacillin
1.5
wk
amoxicillin
v
9
mo
amoxicillin
v
v
clindamycin
iv
clindamycin
iv
v
v
v
9
mo
v
v
v
v
v
penicillin **
v
v
v
v
*
shortly before the next injection was planned; ** plus valaciclovir prophylaxis; wk = week; mo = month; t.i.d. = three times a day;
q.i.d = 4 times a day.
be administered, which may indicate insufficient levels of
the antibiotic during the last phase of the administration
interval. Indeed, in patient B injections every two weeks
and in patient H every three weeks prevented relapses.
prophylaxis has gone unnoticed, because cases are rare.
A survey in our tertiary care centre among 117 cases of
erysipelas yielded eight such cases. It is not usually possible
to judge whether these recurrent episodes represent
relapses from foci within the body (e.g., within the
lymphatic system) or exogenous reinfections.
The analysis of the reasons for failure of preventive therapy
in our sample indicates that the recurrence of erysipelas had
multiple causes: 1) noncompliance; 2) incorrect antibiotic;
3) other causative micro-organism; 4) insufficient antibiotic
concentrations. Importantly, in half of the cases, no valid
explanation could be obtained. Thus, the reasons for failure
of prophylaxis in these cases remain unclear. We tend to
conclude that in such cases, erysipelas may recur despite
concentrations of antibiotics that are otherwise considered
adequate. We have not been able to find many similar cases
in the literature.
No explanation
Recurrences of erysipelas during antibiotic prophylaxis in
patients E, F, and G could not be explained. Furthermore,
no explanation was available for one episode of erysipelas
in patients C and D.
DIS C USSIO N
Our review of the literature and analyses of case reports
clearly indicate that despite antibiotic prophylaxis,
erysipelas still recurs.7,8 Recurrent erysipelas despite
Koster, et al. Recurrent erysipelas despite antibiotic prophylaxis.
MARCH 2007, Vol. 65, No. 3
92
illnesses,10-12 should be considered. Secondly, the clinician
should make sure that the correct antibiotic is selected,
based on the most likely causative micro-organism. Thirdly,
dosing and timing of antibiotic prophylaxis is of great
relevance. A problem is that the penicillin concentrations
needed for adequate prophylaxis are not known. From
a theoretical point of view, it could be argued that the
protective serum penicillin concentrations need to be
maintained at levels equal or above the minimal inhibitory
concentration (MIC) of the causative micro-organism. It is
noteworthy that the majority of the patients in this study
received benzathine penicillin 1.2 MU every four weeks
as the first prophylactic regimen. With this schedule
extremely low penicillin concentrations may be present at
the end of the dosing interval.13 From the literature and also
the cases presented here, it is suggested that three-weekly
schedules are more effective for preventing erysipelas. In
some cases, even two-weekly schedules may be necessary.
Another approach to provide protective plasma penicillin
levels is increasing the dosage.14 Doubling the dose
prolongs the protective plasma penicillin levels by only one
half-life, which in this case may be around seven days.
In addition, the deposition of benzathine penicillin after an
injection in the buttocks has been questioned: the majority
of injections have been reported to be intralipomatous
rather than intramuscular.15 Finally, the clinician should
address the issue of compliance with the prophylactic
treatment. If oral prophylaxis is being prescribed,
patient information is a crucial issue. The importance of
prophylactic treatment to prevent further damage of the
lymph vessels and serious infections should be stressed.
Importantly, in half of the cases no valid explanation could
be obtained. Thus, the reasons for failure of prophylaxis in
the cases remain unclear. The insight that prophylaxis does
not allow the prevention of all episodes of erysipelas may
lead to more systematic investigations of this topic.
An important question is whether evidence exists that
antibiotic prophylaxis is effective in preventing recurrences
of erysipelas. In the older literature, case reports suggested
that prolonged antibiotic prophylaxis successfully reduces
the frequency of attacks in all patients with recurrent
erysipelas.8,9 In addition, several clinical studies have
addressed this problem. In the study by Duvanel et al.5
benzathine penicillin 2.4 MU every three weeks was given
intramuscularly to 12 patients for six months, after the first
attack of erysipelas. There where no recurrences during
the period of prophylaxis, but after discontinuation of
prophylaxis, three patients experienced a recurrence.
Jorup-Rönström et al. 4 included 143 patients with
erysipelas; 29% had recurrences during a follow-up
period of two to four years. Only nine patients received
antibiotic prophylaxis, which prevented further episodes
of erysipelas. After the second recurrence, the calculated
cost of prophylaxis with phenoxymethylpenicillin or
erythromycin was only marginally lower then the cost of
therapy for erysipelas attacks.
Kremer et al.6 studied erythromycin as a preventive
antibiotic in 32 patients who had suffered two or more
episodes of erysipelas or cellulitis during a follow-up period
of 18 months. Only 11 of 36 patients were included because
of recurrent erysipelas. There were no recurrences in the
study group, compared with a relapse rate of 50% in the
control group.
Sjöblom et al.7 prescribed antibiotic prophylaxis to 20 patients
with a history of two or more attacks of erysipelas.
Phenoxymethylpenicillin was given in most cases. A few
patients received erythromycin because of a known allergy
to penicillin. Despite the antibiotic prophylaxis, two patients
had a recurrence, compared with eight patients in the control
group. The median follow-up period was 15 months.
Our study has a number of implications for clinicians
dealing with recurrent erysipelas (summarised in table 3).
First, we would recommend that the diagnosis is as certain
as it can possibly be. Cultures and serology may be of
help to ascertain the cause. Not only should the question
be raised whether it is a streptococcal or a staphylococcal
infection, but also more rare causes, such as Campylobacter
jejuni, especially in patients with compromised host
defences (e.g., hypogammaglobulinaemia) and
Campylobacter fetus in patients with other underlying
REFERENCES
1. Mandell GL, Douglas RG, Bennett JE, Dolin R. Principles and practice of
infectious diseases. 5th ed. Philadelphia: Churchill Livingstone. 2000.
2. Bisno AL, Stevens DL. Streptococcal infections of skin and soft tissues.
N Engl J Med 1996;334:240-5.
3. Dupuy A, Benchikhi H, Roujeau JC, et al. Risk factors for erysipelas of the leg
(cellulits): case-control study. BMJ 1999;318:1591-4.
4. Jorup-Rönström C, Britton S. Recurrent erysipelas: predisposing factors
and costs of prophylaxis. Infection 1987;15:105-6.
Table 3. Recurrent erysipelas despite antibiotic
prophylaxis
5. Duvanel T, Mérot Y, Harms M, et al. Prophylactic antibiotics in erysipelas.
Lancet 1985; 1:1401.
1. Check compliance
2.Reconsider diagnosis (Staphylococcus aureus, Campylobacter
species)
3. Consider shortening the dosing interval (benzathine penicillin)
or raising the dose of oral prophylaxis
4.Change regimen, e.g., to clindamycin
6. Kremer M, Zuckerman R, Avraham Z, et al. Long-term antimicrobial
therapy in the prevention of recurrent soft-tissue infections. J Infection
1991;22:37-40.
7. Sjöblom AC, Eriksson B, Jorup-Rönström C, et al. Antibiotic prophylaxis
in recurrent erysipelas. Infection 1993;21:390-3.
Koster, et al. Recurrent erysipelas despite antibiotic prophylaxis.
MARCH 2007, Vol. 65, No. 3
93
8. Bitnun S. Prophylactic antibiotics in recurrent erysipelas. Lancet
1985;1:345.
12. Monselise A, Blickstein D, Ostfeld I, Segal R, Weinberger M. A case
of cellulitis complication Campylobacter jejuni subspecies jejuni
bacteremia and review of the literature. Eur J Clin Microbiol Infect Dis
2004;23:718-21.
9. Thind P. Prophylactic antibiotics in recurrent erysipelas. Lancet
1985;1:986.
13. Kaplan EL, Berrios X, Speth J, et al. Pharmacokinetics of benzathine
penicillin G: Serum levels during the 28 days after intramuscular injection
of 1,200,000 units. J Pediatr 1989;115:146-50.
10. Kerstens PJ, Endtz HP, Meis JF, et al. Erysipelas-like skin lesions associated
with Campylobacter jejuni septicemia in patients with hypogamma­
globulinemia. Eur J Clin Microbiol Infect Dis 1992;11:842-7.
12. Currie BJ, Burt T, Kaplan EL. Penicillin concentrations after increased
doses of benzathine penicillin G for prevention of secondary rheumatic
fever. Antimicrob Agents Chemother 1994;38:1203-4.
11. Cone LA, Dreisbach PB, Hirschberg J, Secar C, Dreisbach LP, Salatich
W. Cellulitis and Septic Arthritis Caused by Campylobacter fetus and
Campylobacter jejuni: Report of 2 Cases and Review of the Literature.
J Clin Rheumatol 2003;9:362-9.
13. Cockshott WP, Thompson GT, Howlett LJ, Seeley ET. Intramuscular or
intralipomatous injections? N Engl J Med 1982;307:356-8.
Koster, et al. Recurrent erysipelas despite antibiotic prophylaxis.
MARCH 2007, Vol. 65, No. 3
94